Safety and Immunogenicity of Pertussis Vaccine Immunization during Pregnancy: A Meta-Analysis of Randomized Clinical Trials

The objective of this meta-analysis is to assess the safety and immunogenicity of maternal pertussis vaccination based on randomized clinical trials. PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Internet, and Wan Fang Database were searched from inception up to the 8th of October 2021, using a protocol registered on PROSPERO with no. 42021287717, and a meta-analysis was conducted. We measured pooled geometric mean concentrations (GMCs) for IgG antibodies against pertussis and the incidence of serious adverse events (SAEs). We identified a total of 522 publications, and after a strict screening, we found that 6 RCTs were eligible for our meta-analysis. GMCs were determined with a standardized mean difference (SMD), and the pooled SMD of anti-PT, anti-FHA, and anti-PRN IgG from cord blood were 0.91 (95% CI: 0.58, 1.24), 1.03 (95% CI: (0.70, 1.35)), and 1.55(95% CI: 1.22, 1.88), respectively. The pooled OR of SAEs of women and infants did not show a statistical difference; the pooled ORs were 1.26 (95% CI: 0.78, 2.05); P = 0.35) and 0.61 (95% CI: 0.37, 1.01); p = 0.053), respectively. Infants of immunized women have significantly higher transplacental antibodies for protection against pertussis disease during the first 2 months of life.


Introduction
Pertussis is a highly contagious infectious respiratory disease, mainly caused by the bacterium Bordetella pertussis, and typically characterized by a prolonged cough [1]. Although pertussis is vaccine-preventable, it remains a global public health concern [2]. According to a recent report [3], it was estimated that there were 24.1 million cases of pertussis around the world in children aged <5 years with 160,700 deaths and many hospitalization admissions, some to pediatric intensive care units in 2014. Young infants < 6 months of age are at increased risk of pertussis-related complications, and infants <2 months of age are more likely to be under the threat of severe and potentially lethal complications [4]. Besides, pertussis continued to represent a serious public health problem in many countries, even in those with high rates of vaccination coverage [5].
Pertussis vaccination of pregnant women was frst recommended in the United States (US) and the United Kingdom (UK) nearly a decade ago, in response to the resurgence of pertussis disease in the general population and multiple deaths in infants [6,7]. Te frst routine pertussis vaccination occurs at six weeks to three months of age [8]. Infants under two months of age are the most vulnerable and have the highest rate of serious clinical complications requiring hospitalization and the highest mortality rate [9]. Meanwhile, maternal immunization is increasingly being recommended as a strategy to protect young infants from infectious diseases [10]. It was also reported that vaccination during pregnancy results in high levels of antibodies in the mother and the newborn. Furthermore, maternal tetanus, diphtheria, and acellular pertussis (Tdap) vaccination ofers protection for neonates against clinical pertussis until primary vaccinations. Te pertussis vaccine exists in both whole-cell (Tdwp) and acellular (Tdap) forms. Te Tdap form has fewer adverse efects and seems to be as efective as the Tdwp formulation. As a result, the Tdwp preparation is only recommended when the Tdap form is not available [11]. Trough transplacental transfer, antipertussis antibodies pass to the fetus, which is protected at the time of birth and during the frst months of life [12]. However, performing clinical trials in pregnant women is challenging [13]; hence, the vast majority of immunogenicity and safety data has come from observational studies, which are prone to bias [14].
So far, several systematic reviews have investigated the efectiveness and/or safety of pertussis vaccination during pregnancy [15][16][17][18][19][20]. And yet none of them were specifcally conducted as randomized clinical trials (RCT) or addressed the quantitative immune response comprising safety as well as immunogenicity for mother and child. Terefore, we performed a meta-analysis of RCTs to compare the immunogenicity and safety of pertussis vaccination during pregnancy.

Data Sources and Search Strategy.
Tis meta-analysis was conducted according to the Cochrane Collaboration guidelines and preferred reporting items for systematic reviews and meta-analyses protocols (PRISMA protocol) [21] and prospectively registered with PROSPERO (CRD-42021287717). We conducted a systematic search in electronic databases, including PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Internet, and Wan Fang Database, from inception up to the 8th of October, 2021, without language restrictions. Te search strategy was built based on the following keywords and MeSH terms: "maternal," "pregnancy," "pregnant," "pertussis," "vaccination," "vaccine," "randomized controlled trials," and fltered to "clinical trials" and "randomized controlled trials."

Inclusion and Exclusion Criteria.
Articles that met the following criteria were included: (a) RCTs; (b) primary studies; (c) the experimental group was treated with the Tdap vaccine during pregnancy; (d) control groups were treated with either a placebo, standard vaccination, or were unvaccinated; (e) reports at least one immunological response to vaccination. We excluded the following: (a) articles irrelevant to the topic; (b) duplicate publications; (c) trials of a cross-over study design; (d) animal and laboratory studies.

Quality Assessment.
Te methodological quality of each trial was evaluated for risk of bias using standard criteria: method of randomization; allocation concealment; patient, investigator, and outcome assessor blinding; selective outcome reporting; incomplete outcome ascertainment; and other potential sources of bias as recommended by the Cochrane Collaboration [22]. Each domain was categorized as low, high, or unclear.

Data Extraction.
Te data were carefully evaluated and extracted independently from all the eligible publications. Te following data were collected from each study: (a) name of the frst author, year of publication, and geographic setting; (b) study design; (c) type of vaccine during pregnancy; (d) study period; (e) the number of subjects in each group; (f ) registration number of the trial; (g) gestational age in weeks of vaccination. To evaluate maternal pertussis vaccine immunogenicity, geometric mean concentrations (GMCs) for IgG antibodies against pertussis toxin (PT), flamentous haemagglutinin (FHA), and pertactin (PRN) in infants for all vaccine antigens were extracted from the trials. Te following outcomes were considered for the metaanalysis: GMC after the infant series (at delivery, before primary vaccination, and after primary vaccination) of the Tdap vaccine. To evaluate safety, we measured the incidence of serious adverse events (SAEs) for women and their infants.

Statistical
Analysis. Te analysis of the immune response was performed mainly on cohorts according to protocol. Calculations of the GMCs of IgG antibodies against PT, FHA, and PRN were performed by taking the anti-ln of the means of the concentration transformations, and the GMCs were determined with the standardized mean diference (SMD). GMCs for antibodies against each vaccine component were calculated with 95% confdence intervals (CIs) in each study, and p values of less than 0.05 are signifcant. Statistical heterogeneity was assessed using the Cochran Q and I 2 measures; an I 2 value above 25% may be considered low heterogeneity, and a value above 50% and 75% were predefned as moderate and high heterogeneity, respectively [23]. Egger's test and Begg's test were conducted to explore the possibility of publication bias for the primary outcome [24,25]. We also planned a priority to perform a leave-one-out sensitivity analysis to ascertain that the estimates were not driven by single trials. STATA, version 15.1 (StataCorp LP, College Station, TX, USA), was used for meta-analysis.

Search Result.
We identifed a total of 522 publications, of which 349 were excluded due to duplication. Screening of titles and abstracts and inclusion criteria led to the exclusion of 151 publications. Of the remaining 22 studies, 6 were found to match our inclusion criteria (see fowchart in Figure 1).

Study Characteristics.
Altogether, 6 randomized controlled trials were included in our fnal quantitative analysis [26][27][28][29][30][31]. Healthy pregnant women 18-45 years old who were not at known risk of pregnancy-related complications and had a normal singleton pregnancy were included in all these six studies. Te total number of enrolled women was 709 in the experimental (Tdap) group and 691 in the control group. Te basic information for the included studies is shown in Table 1. Participants were vaccinated either with the Tdap vaccine (experimental group) or with a placebo/TT/Td/ unvaccinated (control group). Te gestational age in weeks of vaccination was between 18 weeks and 36 weeks, the details of the RCTs identifed in this study are shown in Table 1.

Quality Assessment.
Most of the included studies had low biases, as shown by our quality assessment using the Cochrane assessment tool. Te detailed quality assessment of each included study is shown in Supplementary File 1.

Meta-Analysis of Immunogenicity.
Five studies were included in the analysis of anti-PT and anti-PRN IgG GMCs of infants from cord blood, and four studies reported the related GMCs for the FHA. Te pooled SMD of anti-PT IgG from cord blood was 0.91 (95% CI: 0.58, 1.24; P < 0.0001). Te pooled SMD of GMC for anti-FHA from cord blood was 1.03 (95% CI: 0.70, 1.35; P < 0.00001). Also, the pooled SMD of anti-PRN IgG from cord blood was 1.55 (95% CI: 1.22, 1.88; P < 0.00001). A random-efects model was employed due to the signifcant heterogeneity between diferent antibody responses among these studies (I 2 � 80.5%, 73.7%, and 77.2%, respectively) ( Figure 2). We removed the openlabel trial of Barug et al. [26] for each analysis, and the heterogeneity of anti-PT and anti-PRN both sharply dropped to zero, and the I 2 value of anti-FHA dropped to 41.5%, which fully explained that the high risk of bias was caused by the study design. Te exclusion of this trial from the meta-analysis did not change the overall conclusion. Subsequently, we conducted meta-analyses according to diferent time points of Tdap vaccination (before and after the primary vaccination) of infants towards anti-PT, anti-FHA, and anti-PRN. Because the relevant antibodies in Perrett et al.'s study [30] were only measured in cord blood, they were not included in this part. Te results are listed in Table 2, and related forest plots are shown in Supplementary File 2. Te results suggest that GMCs for pertussis antibodies were higher in the Tdap group than those in the control group.
At delivery and before primary vaccination with a signifcant diference.After primary vaccination, there was signifcantly less anti-FHA antibody among the Tdap group when compared to the control group. However, no significant diference was noticed between anti-PT and anti-PRN antibodies. Sensitivity analyses showed that pooled SMDs did not change after removing any single study, indicating the stability of our results.

Discussion
To the best of our knowledge, this is the frst meta-analysis exploring the immunogenicity and safety of maternal pertussis vaccination based on RCTs. Tis meta-analysis synthesized evidence about the immunogenicity and safety of Tdap vaccination during pregnancy in 6 studies involving more than 1400 pregnant women and infants. We used a systematic strategy and broad search terms in multiple databases to identify as many published clinical trials as possible. Maternal pertussis immunization has undergone a paradigm shift in recent years as evidence emerges of robust efectiveness and safety in protecting young infants and their mothers against pertussis [32]. GMCs against pertussis were assessed by performing an ln transformation,   Journal of Tropical Medicine to get a more intuitional understanding of the immunogenicity of vaccines. For immunogenicity, our results from the analysis of 6 RCTs suggested that GMCs of anti-PT, anti-FHA, and anti-PRN were higher in the Tdap group than the control group at delivery and before primary vaccination of infants, which is consistent with the included studies [26][27][28][29][30][31]. However, after primary vaccination, anti-PT and anti-PRN did not show statistical diferences between the Tdap group and the control group, and GMCs of anti-FHA were statistically less in the Tdap group than the control group, suggesting that maternal immunization with Tdap resulted in high concentrations of pertussis antibodies in infants during the frst 2 months of life until they get primary vaccinated. Tis supports the recommendation of Tdap vaccination during pregnancy to prevent early-infant pertussis disease.
For safety, signifcant diferences were demonstrated in the comparisons of the incidence of serious SAEs, which mainly included pregnancy-induced hypertension, pancreatitis, acute appendicitis, fetal distress resulting in a Csection, congestive heart failure, and gastroenteritis. Both SAEs of pregnant women and their infants showed no signifcant diferences. According to the included studies, none of the SAEs in women and their infants were judged to be attributable to the Tdap vaccine, except that four of these pregnancy-related SAEs were assessed as possible vaccine-  Journal of Tropical Medicine related (preeclampsia, premature delivery, and HELLP syndrome (hemolysis, elevated liver enzymes, low platelet count) in 1 Td recipient and gestational hypertension in 1 Tdap recipient) [31]. Hoang et al.'s study [28] reported 7 SAEs but did not reveal the distribution of the incidences, so we did not include this in the meta-analysis. Other included studies also reported the incidence of non-SAEs [26,28,30], mainly redness and mild local pain, but they were either without signifcant diferences between the Tdap group and control group or without eligible data for pooled analysis; therefore, we did not perform a meta-analysis about non-SAEs. Overall, our results of the Tdap vaccine's safety are consistent with the included studies.
Infants are specifcally prone to bradycardia, hypotension, and cardiac arrest from pertussis. Te development of pulmonary hypertension has been increasingly recognized as a factor contributing to infant mortality at an early age, as it may lead to worsening systemic hypotension and hypoxia [33]. Some of the included studies [27,30,31] reported the occurrence of obstetric or fetal complications in the Tdap and control groups; however, there were no signifcant diferences between them, and the reported data were not eligible for meta-analysis; thus, we did not perform meta-analysis based on obstetric or fetal complications.
Several limitations of the present study must be acknowledged. First, we searched only six databases, and some unpublished studies or publications in other databases may not have been identifed. Second, only a limited number of published RCTs directly compare the immunogenicity and safety of Tdap maternal vaccination. While RCTs are desirable for addressing the impacts of antenatal vaccination timing on vaccine immunogenicity, there are limitations on study design due to the ethical issues raised by delaying vaccination. Tird, there was signifcant heterogeneity among the studies that evaluated GMCs of pertussis antibodies in cord blood and after primary vaccination. Te result of the sensitivity analyses that were performed indicated that the possible reason for the heterogeneity was a diferent trial design; however, the overall conclusions were  not changed. Fourth, the infants in the included study did not get the postvaccine at the same age, with a one-month delay [26]; hence, the corresponding results for GMCs might include bias. Additionally, the doses of the Tdap vaccine have slight diferences among each included study, and the intervention of the control group was diferent as well, which may infuence the fnal result. Finally, fndings from the studies included in this paper are not necessarily applicable to infants and children globally. Follow-up periods were often of necessity short, mostly to less than 12 months of age.
Most of the studies were conducted in communities with many years of use of pertussis vaccines.

Conclusion
Tis meta-analysis shows signifcant evidence that infants of immunized women had signifcantly higher transplacental antibodies for protection against pertussis disease during the vulnerable newborn period before they received their primary immunizations. We analyzed the incidences of SAEs in women and infants as well, and our results support the recommendation for routine Tdap immunization in pregnancy to improve the protection of infants against pertussis disease before primary infant immunization.

Data Availability
Data are included within the article.

Conflicts of Interest
Te authors declare that there are no conficts of interest.